Great little book for getting the information architecture right.

#Shaping categories and labels

Categorizing content is a process of criteria matching, which means we have to answer two questions:

  1. What are the criteria for the category?
  2. Does the content match the criteria?

The first question starts with the macro view: we need to know the big picture, to understand how all the pieces fit together. Once we’ve defined a category—not just identified it, but established its parameters—we can assess content against that definition, answering the second question.

#Four key factors to shape categorization:

  1. The needs of the users
  2. The goals of the business
  3. The current state of the content
  4. The strategic future of the content

#The needs of the user

By the time you’re building categories for web content, you should have already done the research and discovery work that answers your questions about who your users are.

Organize around the customer means organizing around the task that they want to complete.

#The goals of the business

  • What does the business want to accomplish?
  • Consider the narrative and tell stories when you’re building up sections throughout your site.
  • Org chart navigation is a cliché in the industry because it’s such a poor solution.
  • Users don’t care how your business teams function. They just want to complete their tasks.
  • Avoid Conway’s Law.

#The current state of the content

  • Pay attention to the current state of content. It’s a good yardstick to predict future content. The lessons from the audit informs your categories.
  • Use boardthing.com or Trello to visualize the current sitemap.
  • CHECK: boardthing.com

#The LATCH framework for information architecture

Defined by Richard Saul Wurman, proposed that there are five possible methods for organizing anything and everything: Location, alphabet, time, category, and hierarchy.

  • Location. Country could be a way to to organize web content by geographic location. Products organized by room. Mapping to the visitors’ mental models of homes. Think like IKEA is mapping their products to Living Room, Kitchen, Bathroom, etc.
  • Alphabet. Almost as soon as we learn the alphabet, we’re taught to start structuring information by it. But you rarely see A-Z navigation menus on the web. Alphabetizing isn’t for discovery—it’s for research. Make sure to only use alphabetical organization when you’re certain your users know what they’re looking for.
  • Time. Time adds meaning. When a piece of content was published changes how it’s understood.
  • Category. Categorical organization is often more flexible and robust than location or alphabet. It’s excellent for breaking down large data sets into smaller, more usable, more findable pieces. It makes it perfect for discovery. Just don’t fall into the trap of believing categories are the final word. Other elements of design, functionality, and strategy have to help make categories useful.
  • Hierarchy. Hierarchy is about arranging items according to value: from least to most or most to least. The key to understand this method is that it requires value to be assigned to the information. On a social platform a list of “Most Popular” content is typical a hierarchical selection based on analytics. Don’t confuse this with the concepts of page hierarchy or information hierarchy. The former deals with how web pages are arranged (sitemaps) and the latter is consideration for the relative importance of information in a given environment.

Remember to differentiate between methods of grouping and methods of ordering. Location, Alphabet, and Time are all ordering methods. But Category is a grouping method.

#Preparing a content audit

Here are the starter questions to ask yourself:

  • What’s prompting the audit? Are you starting a redesign? Preparing for a migration? Updating brand standards?
  • What are you trying to learn? “Number of pages” is a good start, but you may also want to know about content distribution, site structure, URL patterns, content types, etc.
  • Who needs to use the results of the audit (besides you)? Is a coworker depending on the data to inform their sitemap, their templates, their migration plans? Will it be used to sell or justify additional work?
  • What kind of resources are available for the audit? Do you have time to work on it (even if you’re just barely eking it out)? Would automated tools or additional workers help you?

#Getting started with content audits

Audits impose order on chaos. Audits are your way in to understand the scope of the content on a website. Audits help you define scopes, target vulnerabilities, and identify strengths. Be sure to collect the right data for the right reasons. An audit is any critical review of a website.

There is a difference between content audits and a content inventory. An audit is a process. An inventory is a product. An audit is the action of reviewing a website. An inventory is the artifact that results from the audit.

An early audit in the project is an excellent way to scope the content and structural needs of a website, and the project itself.

Whoever is running the project should know what digital properties are involved. It pays to be aware of the other entities in a company’s content ecosystem.

During the audit you’re just looking for the broad strokes, a sense of what you might need to dig into later.

Try to answer the following:

#What kind of content is it?

Try to get a sense of some of the different content types or styles that currently exist. On an ecommerce site, you can expect product descriptions, support forums, and company news. A university will likely have a mix of marketing content for prospective students and evergreen content for current students. A nonprofit might have a lot of advocacy content and financial documentation.

You’re looking for patterns, such as:

  • How much content is evergreen versus dynamic?
  • Is the content focused on marketing? Is it selling a product or service?
  • Are there articles, blog posts, interviews, or other journalistic writing (what I term storytelling content)?
  • Is the content research-based? Are there white papers and reports? How is it distributed or made available to users?
  • Is there a lot of documentation or support content? How is it distributed or made available to users?
  • Are there forums, bulletin boards, or other community-based content areas?
  • What kind of rich media (images, videos, slideshows, social media integration, etc.) does the site have?

#How is the site structured

You’re not conducting an in-depth structural audit yet, but pay attention to structural clues throughout the site:

  • Is it easy to navigate, or do you get lost quickly?
  • Do the URLs match the expected paths?
  • Are there breadcrumbs and other wayfinding features?
  • Are any areas duplicated or missing from the menus?
  • Can you easily understand the relationships between pages?
  • Are pages with similar content constructed the same way?
  • Does content flow logically on each page?
  • Do things generally work the way you expect them to?

If you’re seeing a lot of inconsistency in the structures and navigation mechanisms, that’s a sign that your project will require more heavy lifting around the information architecture.

#How effective is the content

Gather a sense for how effective the content is. You want a representative sample of high-profile landing pages and pages with heavy stakeholder investment. Is it easy to read? Is it full of typos? Does it sound “on brand”?

#Layer different types of audits

It could be a good idea to run multiple interconnected audits over the course of a project:

  1. A simple audit to scope the website.
  2. Then an automated audit combined with manual scrutiny to inform strategy recommendations.
  3. And finally, a structural audit to lay the foundation for a new sitemap.

Each audit builds on the one that came before it.

#Determining content scope

When auditing for scope, try to answer several questions:

  1. How much content is on the site?
  2. What kind of content is it?
  3. How is the site structured?
  4. How effective is the content?
  5. How is the content managed?

You don’t need detailed answers—you just need to make an educated guess about the type and amount of structural work to come. Still, these are big, bad questions that you need to unpack.